It has become clear that spatio-temporal patterning of neuronal activity reflects a complex interaction between dynamical properties of neurons and those of the networks they form. The constant reorganization of these properties underlies most cognitive processes in the brain, and their dysregulation causes brain pathologies. The development of multisite optical imaging and electrophysiological recording techniques has enabled the identification and monitoring of dynamic network organization (so called functional network structure) on different time and spatial scales. To fully understand functional dynamics among neurons in in vitro and in vivo situations, it is imperative to develop analytical and computational tools to detect and characterize these distributed functional network structures from experimental recordings. Within the theoretical community, much interest has been focused on developing tools that allow detection of community structure in networks. These tools are generally optimized to analyze the physical (i.e., anatomical) space of network connections, and to parse the network connectivity structure into communities. Elucidation of functional connectivity in neuronal networks presents a very specific challenge, since anatomical connectivity is only one of the factors that mediate formation of functional interactions. The primary challenge is to construct tools that efficiently detect the formation of dynamic functional communities based on the spatio- temporal patterning of neural electrical activity, and that reliably quantify the properties of the detected communities. We have recently developed a functional community detection method that answers this challenge. We propose to couple it with optimal metrics that permit robust detection of dynamically changing network communities and test it thoroughly in computational, in vitro and in vivo settings. Specifically, w propose to: develop and test, through computer simulations, a set of linear and non-linear metrics tailored for the measurement of dynamic changes in functional network communities (AIM 1); use these tools to investigate formation of functional communities in dissociated, mouse hippocampal cell cultures (AIM 2); and apply the tools to in vivo hippocampal multisite recordings obtained from freely behaving mice undergoing a cognitive task (AIM 3). Tools to detect dynamic formation of functional network structures from temporal activity patterns in a subset of network elements will have a very significant impact on neuroscience, as well as in other biosciences where such information is available. Within neuroscience, such tools will provide a better understanding of brain function during different cognitive tasks. They will also provide a valuable diagnostic method for identifying functional network pathologies.